FAQ  •  Register  •  Login

Where did my cells go???

<<

robs

Contributor

Posts: 40

Joined: Mon Dec 02, 2013 3:42 pm

Location: University of Connecticut

Post Mon Jan 27, 2014 8:27 pm

Re: Where did my cells go???

Hi Guys,
I was using a CyTOF2, software version 6.0.449.

I updated and tried various concatenation programs with the same two files so I can give you the most up to date info.
Concatenating with FlowJo v7 gives those display issues, not sure if v9 is the same because I'm on PC.
Concatenation done in the newest beta version of FlowJo vX looks fine, although the event length seems stretched when I loaded the file into some programs.

The Cytobank tool and the new concatenation tool in the DVS software (v626) also generated .fcs that matched the original files. They also both preserved the Time data for the individual samples, which was nice. The CyTOF software tool took longer to concatenate, though, and the resulting file was much bigger (196MB) than the files generated with the other programs (all around 70MB). The original .fcs added up to about 200MB so most of the programs actually compressed the files some.

The Cytobank tool can also export the file number as a parameter so you can gate the samples individually, FlowJo v10 sort of does this.

All but the file generated in FlowJo v7 looked great in Cytobank.

So for now, I still like the cytobank tool the best although the others (minus FlowJo 7) seemed okay.

If you want to see that problem file from v7 I can upload it
-Rob
<<

Ofir

Master

Posts: 75

Joined: Thu Nov 07, 2013 12:46 pm

Location: US, CA

Post Sun Feb 02, 2014 3:01 pm

Re: Where did my cells go???

Recently we started injecting more cells to the CyTOF to get more events for each sample.
We set the acquisition time to be pretty long (say 15 minutes), inject 500 uL of cells, wait for it to go through (about 5 min), then follow up with the next injection from the same sample.
This approach effectively generates more events per sample. Although one can easily tell by looking at the time parameter when the second half of the sample was injected, we see no difference between the data form the two injections (from the same source).

This approach was especially useful when using cells from a very "sticky" source such as solid tumors where concentrating the sample resulted in sample line clogs.

Just FYI.
<<

robs

Contributor

Posts: 40

Joined: Mon Dec 02, 2013 3:42 pm

Location: University of Connecticut

Post Mon Feb 03, 2014 3:50 pm

Re: Where did my cells go???

Thanks for the tip, Ofir.
Do you just switch the loops with the new valve control button at the top right?
-Rob
<<

Ofir

Master

Posts: 75

Joined: Thu Nov 07, 2013 12:46 pm

Location: US, CA

Post Mon Feb 03, 2014 6:24 pm

Re: Where did my cells go???

We have the v1 CyTOF so we just Load and Inject... Old school ;-)
<<

mleipold

Guru

Posts: 2156

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Tue Feb 04, 2014 4:28 pm

Re: Where did my cells go???

Re multiple injections on the same sample:

I thought this was common knowledge, among v1 users. The Stanford folks have been doing this for years.

Typical acquisition: at a flow rate of 45uL/min, Acquisition delay=20 sec, Detector Stability delay=20sec. Inject the first loop (500uL), which then starts slowing down vent rate around 400sec. At ~450sec, make a second injection by switching the sample loop valve from Inject to Load, squirt in your sample, then switch back to Inject.

Typically, every ~500 sec, make another injection. So, if you have 1.5 mL of sample, inject at 0sec (before you hit "Run"), 450sec, and 950sec.

We typically add in 150-200sec at the end of an acquisition to help minimize carry-over from one sample to the next; this way, it's not always necessary to run Wash solution between each and every sample.

So, as an example for 1.5mL sample:
Injection 1: 0sec
Injection 2: 450sec
Injection 3: 950sec
-Injection 3 would run til ~1500sec; add in 200 sec extra, and you can have Acquisition halt at 1700-1800sec.


-Mike
<<

robs

Contributor

Posts: 40

Joined: Mon Dec 02, 2013 3:42 pm

Location: University of Connecticut

Post Tue Feb 04, 2014 7:10 pm

Re: Where did my cells go???

Thanks Ofir and Mike,
I just tried on our CyTOF2 and it does seem that you can achieve the same result on v2 using the freshly added "manual valve switch" button. Although it looks like other software updates like the automatic loop switch have made that mostly unnecessary for v2. Plotting against Time, the injections are still clearly separated as Ofir observed on v1.
You guys are a great resource!
-Rob
<<

mleipold

Guru

Posts: 2156

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Tue Feb 04, 2014 7:30 pm

Re: Where did my cells go???

I'm glad it's working for v2's as well....I'll have to let the FACS facility people know here at Stanford, with their new v2.

It's always good to see a nice separation between injections (or, minimally, look at Time to work out when one sample is trailing off and you can inject the next bolus). You can usually see a sharp "front" for each injection, followed by a tailing off due to diffusion and interactions with the tubing. Ideally, it would be a perfect square wave bolus, but in the real world, there's a tail to the longer-time side.

You can do this for any size of sample. I know that the Nolan members who do the large barcoding experiments sometimes have samples that run into the several mL of sample volume. This way, you can acquire all of a sample in one (huge) IMD/FCS file, which is more convenient for downstream analysis. Here at the HIMC, our samples are smaller, so we haven't run many over 1.5mL, and I don't think we've run any over 2.5mL.


Mike
Previous

Return to Introduction to mass cytometry

Who is online

Users browsing this forum: No registered users and 1 guest